[Equest-users] Matching to utility provided bill/usage

[Busman, Michael R]

Matt,

Calibrating a baseline model to ACTUAL utility usage requires detective work and forensics to make sure your model reasonably represents the ACTUAL building (construction, systems, internal loads, operating schedules, equipment efficiency, etc.) as much as possible.  The more data you have, the better.  Although eQUEST suggests many good default values, ACTUAL might vary significantly.

Ideally, you will have the benefit of a good building energy audit with the following:

·         Construction verified against building drawings

·         Lighting surveyed on a room-by-room basis, scheduling information from the operating staff, data loggers, after hours observation, or DDC trend logs

·         Misc. equipment surveyed or number of computers, copiers, nameplate data, etc. provided to you.  If you can only get model # and qty., you may have to research the electrical ratings.  Scheduling information similar to lighting also applies.

·         HVAC equipment surveyed not only for nameplate data, but a physical evaluation of the unit - filters, blowers, OA, RA, MA, SA, space temperatures measured, economizer dampers working or linkages disconnected or dampers closed/screwed shut,  air flow measurements or balance report, coils (cleanliness, fins straight/crushed over, air-cooled condensers degraded due to corrosion from salt-air exposure or pollution which reduces cooling efficiency).  You also want to know how equipment is scheduled and the hours of operation.  This could be anecdotal information from the operating staff, DDC trend logs, data loggers, after hours observation.

·         Central plant equipment surveyed and a physical evaluation of the equipment.  These include but are not limited to boiler stack temps, HHW, temps/steam pressure, boiler combustion efficiency test or most recent test report, operating schedules, condition of steam traps, how chillers & boilers are sequenced, chilled & condenser water temps, check chiller approaches for signs of potential condenser fouling or low refrigerant charge.

If you have the benefit of a building audit, you will be in a better position to define your model to reflect ACTUAL as much as possible and adjust inputs based on measurements, observations, or trend log/logger data.  By the way, a good audit is not limited to the items listed above.

Now, if the building is on a different coast and it's not possible to perform an audit, you need to have building drawings and a maintenance & operations contact to obtain anecdotal information from at a minimum.  Ask questions about the items listed above to get as much information as possible.  You'll then have to play detective to fill in some of the holes in the information needed.  For example, if you're told that HVAC equipment is controlled by a programmable thermostat, find out what type.  Unless it's like some of the more recent types that can maintain start/stop times in memory without a battery backup, chances are that it uses batteries.  9 out of 10 times, these are out of sight, out of mind and the batteries are dead and the programmed time schedules have been lost.  Another example is if the customer has an air-cooled chiller or rooftop package unit located in a coastal environment, there are very good chances that the condenser is degraded and the fins might crumble like crackers.  I've measured chillers with a new, off-the-shelf cooling efficiency of 1.15 kW/ton, however, the condensers were so bad, the measured efficiency was as bad as 2.5 kW/ton.

Make sure you are using a weather file appropriate for the building location.  Remember that the weather files are representative of long-term weather and the ACTUAL energy usage data you have might be from an abnormal year.  Before I forget, use Google Earth to look at the building for things like shading from other buildings or trees, get an idea of % glazing, wall and roof colors, or air-cooled HVAC equipment or cooling towers hidden within an enclosure with tight clearances causing the units to be starved for air or recirculation of hot air discharged from the condenser or tower.

Sometimes, I've had to build a model with only knowledge of the square footage and age of the building.  That's when you need to make educated guesses about insulation, lighting power densities, equipment efficiencies, etc. based on the age of the building.

Sorry for the long-winded dissertation, but if you make the appropriate tweaks in your calibration, you should be able to get to ±7.5%-10% of actual consumption, which is generally considered acceptable that the model reasonably describes the building energy usage.

Good luck,

Mike Busman

Michael R. Busman, CEM
Lead Project Engineer II

Chevron Energy Solutions
A Division of Chevron U.S.A., Inc.
145 S. State College Blvd.
Brea, CA  92821
Direct  714-671-3561
Fax     714-671-3438
eFax   866-420-0335 (Include my Full Name followed by "CAI:MHTZ" on Cover Sheet)
Mobile 310-387-2083
mbusman at chevron.com<mailto:mbusman at chevron.com>


From: equest-users-bounces at lists.onebuilding.org [mailto:equest-users-bounces at lists.onebuilding.org] On Behalf Of -Matt
Sent: Wednesday, January 30, 2013 12:02 AM
To: equest-users at lists.onebuilding.org
Subject: [Equest-users] Matching to utility provided bill/usage

First time sending to this forum.

Quick question:
It's great when I am provided the utility usage (kWh and Therms) for a building but my models always start really far off and I often am not sure what to do to correct it. Since I have ACTUAL usage, is there any way to refine the model to match it? I'm guessing it's just too open ended to do it automatically; If I selected the wrong system or insulation or windows... it won't know which to correct.

Again, sorry if this isn't the proper way to submit, I'm not used to such a 'simple' forum :)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/equest-users-onebuilding.org/attachments/20130130/a7e5c6ef/attachment-0002.htm>